The program project proposal represents an integrated attempt to apply "state-of-the-art" methodologies for theory testing with naturally occurring phenomena to the comparative assessment of theories for drug abuse antecedents and consequences. Many important questions regarding the etiology and consequences of drug use and abuse must be studied in naturally occurring contexts. Typical experimental methods that assign subjects randomly to conditions of drug use leading to various degrees of abuse are unethical. Animal models and biochemical methods also cannot address many important questions, such as the psychosocial, financial, and generally disruptive effects of drug use. Nonetheless, it is critical to compare and contrast theories that attempt to explain drug-taking etiologies and consequences. This program project consists of six interrelated research studies that develop, compare, and contrast numerous theoretical models in new and existing databases from psychology, epidemiology, psychiatry, education, and sociology, using modern multivariate psychometric/statistical methods to control for artifacts and competing hypotheses. Under the umbrella of the program project, these studies are both theory-driven and relevant to drug use, yet range from the development of new statistical methods to analyze drug abuse data, to a unique effort to study the dynamics of gang membership. The Models and Methods Project is extending the state-of-the-art in latent variable modeling to a variety of types of data. The Longitudinal Consequences Project tests theories of long-term consequences of drug use through a 20-year follow-up of a large sample of adolescents, and adds a community sample to cross-validate findings. The Collaborative Project works with investigators who have generated over a dozen valuable epidemiological and longitudinal datasets to answer a variety of questions about the etiology and consequences of drug use. The African-American Drug Abuse Project is a longitudinal study contrasting those adolescents at high risk for drug use and abuse with others in their community. The Memory and Cognitive Processes in Drug Abuse Project uses contemporary psychological theories on memory and cognition to improve and test theories of drug use. The Computer Intensive Hypothesis Testing Project uses computer simulation to provide near-exact test performance for standard statistics that behave badly on drug abuse data. The Core Project integrates all components, hosts visiting scholars, and provides free computer access to national drug abuse researchers. It also proposes to provide opportunities for minority high school and undergraduate students to participate in some of its work.